Random walks based segmentation approach for image retrieval

Author

Tabout, Hassan ; CHAHIR, Youssef ; Souissi, Abdelmoghit ; Sbihi, Abderrahmane

Author_Institution

LASTID, Ibntofail Univ.

Volume

2

fYear

2008

fDate

6-8 Oct. 2008

Firstpage

593

Lastpage

597

Abstract

In this paper a region-based image indexing and retrieval (RBIR) algorithm is presented. As a basis for the indexing, a novel spectral segmentation approach using random walks on graphs is introduced. Based on the extracted regions, characteristic features are estimated using color and texture information. The focus of this study is to improve the capture of regions so as to enhance indexing and retrieval performance. An implemented prototype system has demonstrated a promising retrieval performance for a test database containing more than 1000 color algae images.

Keywords

database indexing; feature extraction; graph theory; image colour analysis; image retrieval; image segmentation; image texture; random processes; characteristic feature estimation; color information; graph theory; image retrieval; random walk; region extraction; region-based image indexing; spectral segmentation approach; texture information; Data mining; Feature extraction; Focusing; Image databases; Image retrieval; Image segmentation; Indexing; Information retrieval; Prototypes; System testing; Diffusion; Image retrieval; Markov chain; graphs; random walks;

fLanguage

English

Publisher

ieee

Conference_Titel

Pervasive Computing and Applications, 2008. ICPCA 2008. Third International Conference on

Conference_Location

Alexandria

Print_ISBN

978-1-4244-2020-9

Electronic_ISBN

978-1-4244-2021-6

Type

conf

DOI

10.1109/ICPCA.2008.4783681

Filename

4783681